Romaín Blazevic
I’m a Data Engineer and MLOps Engineer based in Paris, where I design and operate scalable data and ML pipelines across cloud platforms like Azure and AWS.
About
I’m a Data Engineer and MLOps Engineer based in Paris, where I design and operate scalable data and ML pipelines across cloud platforms like Azure and AWS. I build end-to-end architectures that connect ingestion, orchestration, storage, processing, and observability—so teams can trust the data, monitor model behavior, and ship improvements reliably.
I enjoy turning complex requirements into robust systems: streaming and batch ETL, KPI-driven dashboards, alerting, model deployment automation, and infrastructure-as-code with Terraform. My focus is always on production-grade reliability, performance, and maintainability—whether I’m optimizing Spark pipelines on partitioned data lakes, integrating Kafka-based event flows, or setting up real-time monitoring for ML drift and system health.
- 5 years' experience
- Joined October 2026
Skills· 13
Languages
- EnglishAdvanced
- FrenchAdvanced
Industry experience
- Software & Internet
- Financial Services
- Education
- Computers & Electronics
- Other
Work experience· 5 roles
Developed and operated cloud data platform capabilities on AWS. Built an evolving data lake on S3 with structured zone layers (raw, staged, curated) for analytics and ML use cases. Integrated Athena for querying raw/processed data without extra infra. Migrated ETL pipelines from scripts/jobs to AWS Glue to reduce job failures and enable a centralized ML functions/data marketplace style approach; accelerated ML development cycles and enabled dataset reuse. Deployed an enterprise data warehouse on Amazon Redshift to analyze across multiple sources (S3, RDS, DynamoDB), replacing ad-hoc Excel and SQL site databases. Built Spark pipelines using EMR/Glue with partitioning and optimized parallelism, and implemented real-time monitoring/alerting using AWS CloudWatch. Automated Kubernetes pipeline CI/CD using GitLab CI and Argo CD.
Built and optimized data processing and ML/RAG pipelines: designed Python/Kubernetes/Kafka-based data ingestion and processing for artifact/object storage, improved Spark cluster query performance, created a vehicle-fire tracking program, and developed RAG pipelines and LLM configuration running on local hardware. Implemented performance and accuracy improvements for import/process/storage workflows and optimized compression and storage strategies. Developed real-time dashboards/visualization for battery/vehicle telemetry, and handled provisioning/management for compute and material needs.
Deployed and maintained production data/ML services across Azure/AWS. Implemented automated CI/CD for Kubernetes using GitLab CI and Argo CD, built telemetry/metrics collection systems for streaming pipelines with Kafka on Kubernetes, and supported real-time aggregation and performance monitoring. Worked on scalable ML data pipelines and platform components (Python/Spark/ETL, databases like Azure PostgreSQL and MongoDB, and observability with Grafana/Elastic). Created automated deployment workflows for web applications on multi-region serverless-like environments using AWS services and infrastructure as code (Terraform/Ansible).
Education
Certifications and awards
Hire Romaín Blazevic today
Post a job and invite Romaín, or contact them directly. No hidden commission.